2 ms·
> For N balls, the other balls can/should be binary searched. Interesting assertion. Can you prove that? Can you give a provably minimal search pattern? >
by RiderOfGiraffes 17y ago
> For N balls, the other balls can/should be binary searched.
Interesting assertion. Can you prove that? Can you give a provably minimal search pattern?
> It's just a little tricky when it comes to the last two.
Another interesting assertion. It seems to me that the case of exactly two balls is trivial, once the solution is found.
- OmniLarry 17y agoSorry, can't find the link, but it was a Topcoder problem at least 5 years ago. Well, it is trivial with two balls, once you get it, but it's tricky with respect to "just binary search".